[1]ZHAO Wenqing,ZHAO Zhenhuan,GONG Jiaxiao.Remote sensing image object detection based on inverted residual self-attention mechanism[J].CAAI Transactions on Intelligent Systems,2025,20(1):64-72.[doi:10.11992/tis.202312001]
Copy

Remote sensing image object detection based on inverted residual self-attention mechanism

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 20 Number of periods: 2025 1 Page number: 64-72 Column: 学术论文—机器学习 Public date: 2025-01-05
Title:
Remote sensing image object detection based on inverted residual self-attention mechanism
Author(s):
ZHAO Wenqing1; 2;  ZHAO Zhenhuan1;  GONG Jiaxiao1
1. School of Control and Computer Engineering, North China Electric Power University, Baoding 071003, China;
2. Hebei Key Laboratory of Knowledge Computing for Energy & Power, Baoding 071003, China
Keywords:
remote sensing image; object detection; inverted residual; self-attention mechanism; multi-scale; spatial pyramid; feature extraction; feature fusion
CLC:
TP391
DOI:
10.11992/tis.202312001
Abstract:
An inverted residual self-attention method (IRSAM) was proposed in this study as an approach for object detection in remote sensing images. The method was designed to address challenges related to significant variations in object sizes and substantial interference from background information in remote sensing image object detection. Firstly, an inverted residual self-attention mechanism backbone network with strong feature extraction ability was utilized to fully extract the object features, thus reducing the interference of complex background information on the object. Additionally, a multi-scale spatial pyramid pooling module was constructed to offer diverse sensory fields at multiple scales and improve the capacity to detect objects of varying sizes. Finally, a lightweight feature fusion structure was employed to integrate the feature maps extracted from the backbone network, effectively combining low-level and high-level features. The study compared the performance of IRSAM with both traditional network and enhanced object detection algorithms. The results indicated that the proposed method exhibited significantly higher detection accuracy. In addition, ablation experiments were designed on the DIOR and the RSOD datasets. The results show that the mean accuracy is 4.6 and 4.2 percentage points higher than the YOLOv8 algorithm on the DIOR dataset and the RSOD dataset, respectively. Consequently, the proposed method significantly enhances the accuracy of object detection in remote sensing images.
References:
[1] 石争浩, 仵晨伟, 李成建, 等. 航空遥感图像深度学习目标检测技术研究进展[J]. 中国图象图形学报, 2023, 28(9): 2616-2643.
SHI Zhenghao, WU Chenwei, LI Chengjian, et al. Research progress on deep learning object detection technology for aerial remote sensing images[J]. Journal of image and graphics, 2023, 28(9): 2616-2643.
[2] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus: IEEE, 2014: 580-587.
[3] GIRSHICK R. Fast R-CNN[C]//2015 IEEE International Conference on Computer Vision. Santiago: IEEE, 2015: 1440-1448.
[4] REN Shaoqing, HE Kaiming, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 39(6): 1137-1149.
[5] LU Dongchen, LI Dongmei, LI Yali, et al. OSKDet: orientation-sensitive keypoint localization for rotated object detection[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE, 2022: 1172-1182.
[6] XIA Guisong, BAI Xiang, DING Jian, et al. DOTA: a large-scale dataset for object detection in aerial images[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 3974-3983.
[7] LIU Wei, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[M]//Lecture Notes in Computer Science. Cham: Springer International Publishing, 2016: 21-37.
[8] 唐嘉潞, 杨钟亮, 张凇, 等. 结合显微视觉和注意力机制的毛羽检测方法[J]. 智能系统学报, 2022, 17(6): 1209-1219.
TANG Jialu, YANG Zhongliang, ZHANG Song, et al. Detection of yarn hairiness combining microscopic vision and attention mechanism[J]. CAAI transactions on intelligent systems, 2022, 17(6): 1209-1219.
[9] GUO Chaoxu, FAN Bin, ZHANG Qian, et al. AugFPN: improving multi-scale feature learning for object detection[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: IEEE, 2020: 12592-12601.
[10] 吴珺, 董佳明, 刘欣, 等. 注意力优化的轻量目标检测网络及应用[J]. 智能系统学报, 2023, 18(3): 506-516.
WU Jun, DONG Jiaming, LIU Xin, et al. Attention-optimized lightweight object detection network and its application[J]. CAAI transactions on intelligent systems, 2023, 18(3): 506-516.
[11] 胡硕, 王洁, 孙妍, 等. 无人机视角下的多车辆跟踪算法研究[J]. 智能系统学报, 2022, 17(4): 798-805.
HU Shuo, WANG Jie, SUN Yan, et al. Research on multi-vehicle tracking algorithm from the perspective of UAV[J]. CAAI transactions on intelligent systems, 2022, 17(4): 798-805.
[12] FARHADI A, REDMON J. YOLOv3: an incremental improvement[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 1804-2767.
[13] 曾文健, 朱艳, 沈韬, 等. 面向非对称特征注意力和特征融合的太赫兹图像检测[J]. 中国图象图形学报, 2022, 27(8): 2496-2505.
ZENG Wenjian, ZHU Yan, SHEN Tao, et al. Terahertz image detection combining asymmetric feature attention and feature fusion[J]. Journal of image and graphics, 2022, 27(8): 2496-2505.
[14] BOCHKOVSKIY A, WANG C. Y, LIAO H. M, et al. YOLOv4: optimal speed and accuracy of object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Seattle: IEEE, 2020: 2-7.
[15] 张正, 马渝博, 柳长安, 等. 面向遥感图像旋转目标检测的双向衰减损失方法研究[J]. 电子与信息学报, 2023, 45(10): 3578-3586.
ZHANG Zheng, MA Yubo, LIU Chang’an, et al. Research on bidirectional attenuation loss method for rotating object detection in remote sensing image[J]. Journal of electronics & information technology, 2023, 45(10): 3578-3586.
[16] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C]//2017 IEEE International Conference on Computer Vision. Venice: IEEE, 2017: 2999-3007.
[17] 曲海成, 王蒙, 柴蕊. 双向多尺度特征融合的高效遥感图像车辆检测[J]. 计算机工程与应用, 2024, 60(12): 346-356.
QU Haicheng, WANG Meng, CHAI Rui. Efficient remote sensing image vehicle detection with bidirectional multi-scale feature fusion [J]. Computer engineering and applications, 2024, 60(12): 346-356.
[18] SANDLER M, HOWARD A, ZHU Menglong, et al. MobileNetV2: inverted residuals and linear bottlenecks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 4510-4520.
[19] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 770-778.
[20] MEHTA S, RASTEGARI M. Mobilevit: light-weight, general-purpose, and mobile-friendly vision transformer[EB/OL]. (2021-10-05)[2023-11-26]. https://arxiv.org/abs/2110.02178.
[21] ZHANG Jiangning, LI Xiangtai, LI Jian, et al. Rethinking mobile block for efficient attention-based models[C]//2023 IEEE/CVF International Conference on Computer Vision. Paris: IEEE, 2023: 1389-1400.
[22] 郑哲, 雷琳, 孙浩, 等. FAGNet: 基于MAFPN和GVR的遥感图像多尺度目标检测算法[J]. 计算机辅助设计与图形学学报, 2021, 33(6): 883-894.
ZHENG Zhe, LEI Lin, SUN Hao, et al. FAGNet: A multi-scale object detection algorithm for remote sensing images based on MAFPN and GVR[J]. Journal of computer-aided design and computer graphics, 2021, 33(6): 883-894.
[23] 赵文清, 康怿瑾, 赵振兵, 等. 改进YOLOv5s的遥感图像目标检测[J]. 智能系统学报, 2023, 18(1): 86-95.
ZHAO Wenqing, KANG Yijin, ZHAO Zhenbing, et al. Remote sensing image object detection based on improved YOLOv5s[J]. CAAI transactions on intelligent systems, 2023, 18(1): 86-95.
[24] ZHAO Wenqing, KANG Yixin, CHEN Hao, et al. Adaptively attentional feature fusion oriented to multiscale object detection in remote sensing images[J]. IEEE transactions on instrumentation and measurement, 2023, 72: 1-11.
[25] YUE Chenke, YAN Junhua, ZHANG Yin, et al. SCFNet: Semantic correction and focus network for remote sensing image object detection[J]. Expert systems with applications, 2023, 224: 119980.
[26] LI Wentong, CHEN Yijie, HU Kaixuan, et al. Oriented reppoints for aerial object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE, 2022: 1829-1838.
[27] REN Shougang, FANG Zhiruo, GU Xingjian. A cross stage partial network with strengthen matching detector for remote sensing object detection[J]. Remote sensing, 2023, 15(6): 1574.
[28] CHEN Yixia, LIN Mingwei, HE Zhu, et al. Consistency-and dependence-guided knowledge distillation for object detection in remote sensing images[J]. Expert systems with applications, 2023, 229: 120519.
[29] NASCIMENTO M G, FAWCETT R, PRISACARIU V A Dsconv: efficient convolution operator[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019: 5148-5157.
Similar References:

Memo

-

Last Update: 2025-01-05

Copyright © CAAI Transactions on Intelligent Systems